Publications

Abstract not provided.

In November 1988, the US Nuclear Regulatory Commission (NRC) issued Generic Letter 88-20 requesting that all licensees perform an individual Plant Examination (IPE) to identify any plant-specific vulnerability to severe accidents and report the results to the Commission. This paper provides perspectives gained from reviewing 75 Individual Plant Examination (IPE) submittals covering 108 nuclear power plant units. Variability both within and among reactor types is examined to provide perspectives regarding plant-specific design and operational features, and modeling assumptions that play a significant role in the estimates of core damage frequencies in the IPEs.

This paper discusses the core damage frequency (CDF) insights gained by analyzing the results of the Individual Plant Examinations (IPES) for two groups of plants: boiling water reactor (BWR) 3/4 plants with Reactor Core Isolation Cooling systems, and Westinghouse 4-loop plants. Wide variability was observed for the plant CDFs and for the CDFs of the contributing accident classes. On average, transients-with loss of injection, station blackout sequences, and transients with loss of decay heat removal are important contributors for the BWR 3/4 plants, while transients, station blackout sequences, and loss-of-coolant accidents are important for the Westinghouse 4-loop plants. The key factors that contribute to the variability in the results are discussed. The results are often driven by plant-specific design and operational characteristics, but differences in modeling approaches are also important for some accident classes.

An accident management strategy has been proposed in which the reactor coolant system is intentionally depressurized during an accident. The aim is to reduce the containment pressurization that would result from high pressure ejection of molten debris at vessel breach. Probabilistic risk assessment (PRA) methods were used to evaluate this strategy for the Surry nuclear power plant. Sensitivity studies were conducted using event trees that were developed for the NUREG-1150 study. It was found that depressurization (intentional or unintentional) had minimal impact on the containment failure probability at vessel breach for Surry because the containment loads assessed for NUREG-1150 were not a great threat to the containment survivability. An updated evaluation of the impact of intentional depressurization on the probability of having a high pressure melt ejection was then made that reflected analyses that have been performed since NUREG-1150 was completed. The updated evaluation confirmed the sensitivity study conclusions that intentional depressurization has minimal impact on the probability of a high pressure melt ejection. The updated evaluation did show a slight benefit from depressurization because depressurization delayed core melting, which led to a higher probability of recovering emergency core coolant injection, thereby arresting the core damage.

An evaluation of the key elements affecting Direct Containment Heating (DCH) was performed for the Surry plant. This involved determining the dominant high pressure core damage sequences, the probability of proceeding to vessel breach at high pressure, the DCH loads, and the containment strength. Each of these factors was evaluated separately, and then the results were combined to give the overall threat from DCH. The maximum containment failure probability by DCH for Surry is 10{sup {minus}3} when considering four base DCH scenarios and using the two-cell equilibrium (TCE) model. However, higher contamination failure probabilities are estimated in sensitivity cases. When the depressurization and containment loads aspects are combined, the containment failure probability (conditional on station blackout sequence) is less than 19{sup {minus}2}. CONTAIN calculations were performed to provide insights regarding DCH phenomenological uncertainties and potential conservatisms in the TCE model. The CONTAIN calculations indicated that the TCE calculations were conservative for Surry and that the dominant factors were neglect of heat transfer to surroundings and complete combustion of hydrogen on DCH time scales.

A feasibility study for developing an improved tool and improved models for performing event assessments is described. The study indicates that the IRRAS code should become the base tool for performing event assessments, but that modifications would be needed to make it more suitable for routine use. Alternative system modeling approaches are explored and an approach is recommended that is based on improved train-level models. These models are demonstrated for Grand Gulf and Sequoyah. The insights that can be gained from importance measures are also demonstrated. The feasibility of using Individual Plant Examination (IPE) submittals as the basis for train-level models for precursor studies was also examined. The level of reported detail was found to vary widely, but in general, the submittals did not provide sufficient information to fully define the model. The feasibility of developing an industry risk profile from precursor results and of trending precursor results for individual plants were considered. The data sparsity would need to be considered when using the results from these types of evaluations, and because of the extremely sparse data for individual plants we found that trending evaluations for groups of plants would be more meaningful than trending evaluations for individual plants.

Results of calculations performed with MELCOR and HECTR in support of the NUREG-1150 study are presented in this report. The analyses examined a wide range of issues. The analyses included integral calculations covering an entire accident sequence, as well as calculations that addressed specific issues that could affect several accident sequences. The results of the analyses for Grand Gulf, Peach Bottom, LaSalle, and Sequoyah are described, and the major conclusions are summarized.

The use of NUREG-1150 and similar Probabilistic Risk Assessments in NRC and industry risk management programs is discussed. Risk management'' is more comprehensive than the commonly used term accident management.'' Accident management includes strategies to prevent vessel breach, mitigate radionuclide releases from the reactor coolant system, and mitigate radionuclide releases to the environment. Risk management also addresses prevention of accident initiators, prevention of core damage, and implementation of effective emergency response procedures. The methods and results produced in NUREG-1150 provide a framework within which current risk management strategies can be evaluated, and future risk management programs can be developed and assessed. Examples of the use of the NUREG-1150 framework for identifying and evaluating risk management options are presented. All phases of risk management are discussed, with particular attention given to the early phases of accidents. Plans and methods for evaluating accident management strategies that have been identified in the NRC accident management program are discussed. 2 refs., 3 figs.

The MELCOR computer code, which has been developed at Sandia National Laboratories for the US Nuclear Regulatory Commission as a tool for calculating realistic estimates of severe accident consequences and source terms, has been used to analyze a series of containment issues for station blackout sequences for the Grand Gulf Nuclear Power Plant. The results indicate that there is a limited time interval in which the drywell atmosphere would be flammable, and that this would only occur if the vacuum breaker were to stick open within a narrow time window. If burning does occur during this time, it appears quite likely that it would not pose a threat to the drywell wall. The main conclusion from this study is that the drywell atmosphere is not very likely to be flammable for a station blackout sequence. 1 ref. (S.J.)